linux-stable.git/net, branch v4.4.32

packet: on direct_xmit, limit tso and csum to supported devices

2016-11-15T06:46:39+00:00

[ Upstream commit 104ba78c98808ae837d1f63aae58c183db5505df ]

When transmitting on a packet socket with PACKET_VNET_HDR and
PACKET_QDISC_BYPASS, validate device support for features requested
in vnet_hdr.

Drop TSO packets sent to devices that do not support TSO or have the
feature disabled. Note that the latter currently do process those
packets correctly, regardless of not advertising the feature.

Because of SKB_GSO_DODGY, it is not sufficient to test device features
with netif_needs_gso. Full validate_xmit_skb is needed.

Switch to software checksum for non-TSO packets that request checksum
offload if that device feature is unsupported or disabled. Note that
similar to the TSO case, device drivers may perform checksum offload
correctly even when not advertising it.

When switching to software checksum, packets hit skb_checksum_help,
which has two BUG_ON checksum not in linear segment. Packet sockets
always allocate at least up to csum_start + csum_off + 2 as linear.

Tested by running github.com/wdebruij/kerneltools/psock_txring_vnet.c

  ethtool -K eth0 tso off tx on
  psock_txring_vnet -d $dst -s $src -i eth0 -l 2000 -n 1 -q -v
  psock_txring_vnet -d $dst -s $src -i eth0 -l 2000 -n 1 -q -v -N

  ethtool -K eth0 tx off
  psock_txring_vnet -d $dst -s $src -i eth0 -l 1000 -n 1 -q -v -G
  psock_txring_vnet -d $dst -s $src -i eth0 -l 1000 -n 1 -q -v -G -N

v2:
  - add EXPORT_SYMBOL_GPL(validate_xmit_skb_list)

Fixes: d346a3fae3ff ("packet: introduce PACKET_QDISC_BYPASS socket option")
Signed-off-by: Willem de Bruijn 
Acked-by: Eric Dumazet 
Acked-by: Daniel Borkmann 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

sctp: validate chunk len before actually using it

2016-11-15T06:46:39+00:00

[ Upstream commit bf911e985d6bbaa328c20c3e05f4eb03de11fdd6 ]

Andrey Konovalov reported that KASAN detected that SCTP was using a slab
beyond the boundaries. It was caused because when handling out of the
blue packets in function sctp_sf_ootb() it was checking the chunk len
only after already processing the first chunk, validating only for the
2nd and subsequent ones.

The fix is to just move the check upwards so it's also validated for the
1st chunk.

Reported-by: Andrey Konovalov 
Tested-by: Andrey Konovalov 
Signed-off-by: Marcelo Ricardo Leitner 
Reviewed-by: Xin Long 
Acked-by: Neil Horman 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net sched filters: fix notification of filter delete with proper handle

2016-11-15T06:46:39+00:00

[ Upstream commit 9ee7837449b3d6f0fcf9132c6b5e5aaa58cc67d4 ]

Daniel says:

While trying out [1][2], I noticed that tc monitor doesn't show the
correct handle on delete:

$ tc monitor
qdisc clsact ffff: dev eno1 parent ffff:fff1
filter dev eno1 ingress protocol all pref 49152 bpf handle 0x2a [...]
deleted filter dev eno1 ingress protocol all pref 49152 bpf handle 0xf3be0c80

some context to explain the above:
The user identity of any tc filter is represented by a 32-bit
identifier encoded in tcm->tcm_handle. Example 0x2a in the bpf filter
above. A user wishing to delete, get or even modify a specific filter
uses this handle to reference it.
Every classifier is free to provide its own semantics for the 32 bit handle.
Example: classifiers like u32 use schemes like 800:1:801 to describe
the semantics of their filters represented as hash table, bucket and
node ids etc.
Classifiers also have internal per-filter representation which is different
from this externally visible identity. Most classifiers set this
internal representation to be a pointer address (which allows fast retrieval
of said filters in their implementations). This internal representation
is referenced with the "fh" variable in the kernel control code.

When a user successfuly deletes a specific filter, by specifying the correct
tcm->tcm_handle, an event is generated to user space which indicates
which specific filter was deleted.

Before this patch, the "fh" value was sent to user space as the identity.
As an example what is shown in the sample bpf filter delete event above
is 0xf3be0c80. This is infact a 32-bit truncation of 0xffff8807f3be0c80
which happens to be a 64-bit memory address of the internal filter
representation (address of the corresponding filter's struct cls_bpf_prog);

After this patch the appropriate user identifiable handle as encoded
in the originating request tcm->tcm_handle is generated in the event.
One of the cardinal rules of netlink rules is to be able to take an
event (such as a delete in this case) and reflect it back to the
kernel and successfully delete the filter. This patch achieves that.

Note, this issue has existed since the original TC action
infrastructure code patch back in 2004 as found in:
https://git.kernel.org/cgit/linux/kernel/git/history/history.git/commit/

[1] http://patchwork.ozlabs.org/patch/682828/
[2] http://patchwork.ozlabs.org/patch/682829/

Fixes: 4e54c4816bfe ("[NET]: Add tc extensions infrastructure.")
Reported-by: Daniel Borkmann 
Acked-by: Cong Wang 
Signed-off-by: Jamal Hadi Salim 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

udp: fix IP_CHECKSUM handling

2016-11-15T06:46:39+00:00

[ Upstream commit 10df8e6152c6c400a563a673e9956320bfce1871 ]

First bug was added in commit ad6f939ab193 ("ip: Add offset parameter to
ip_cmsg_recv") : Tom missed that ipv4 udp messages could be received on
AF_INET6 socket. ip_cmsg_recv(msg, skb) should have been replaced by
ip_cmsg_recv_offset(msg, skb, sizeof(struct udphdr));

Then commit e6afc8ace6dd ("udp: remove headers from UDP packets before
queueing") forgot to adjust the offsets now UDP headers are pulled
before skb are put in receive queue.

Fixes: ad6f939ab193 ("ip: Add offset parameter to ip_cmsg_recv")
Fixes: e6afc8ace6dd ("udp: remove headers from UDP packets before queueing")
Signed-off-by: Eric Dumazet 
Cc: Sam Kumar 
Cc: Willem de Bruijn 
Tested-by: Willem de Bruijn 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: sctp, forbid negative length

2016-11-15T06:46:38+00:00

[ Upstream commit a4b8e71b05c27bae6bad3bdecddbc6b68a3ad8cf ]

Most of getsockopt handlers in net/sctp/socket.c check len against
sizeof some structure like:
        if (len < sizeof(int))
                return -EINVAL;

On the first look, the check seems to be correct. But since len is int
and sizeof returns size_t, int gets promoted to unsigned size_t too. So
the test returns false for negative lengths. Yes, (-1 < sizeof(long)) is
false.

Fix this in sctp by explicitly checking len < 0 before any getsockopt
handler is called.

Note that sctp_getsockopt_events already handled the negative case.
Since we added the < 0 check elsewhere, this one can be removed.

If not checked, this is the result:
UBSAN: Undefined behaviour in ../mm/page_alloc.c:2722:19
shift exponent 52 is too large for 32-bit type 'int'
CPU: 1 PID: 24535 Comm: syz-executor Not tainted 4.8.1-0-syzkaller #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.9.1-0-gb3ef39f-prebuilt.qemu-project.org 04/01/2014
 0000000000000000 ffff88006d99f2a8 ffffffffb2f7bdea 0000000041b58ab3
 ffffffffb4363c14 ffffffffb2f7bcde ffff88006d99f2d0 ffff88006d99f270
 0000000000000000 0000000000000000 0000000000000034 ffffffffb5096422
Call Trace:
 [] ? __ubsan_handle_shift_out_of_bounds+0x29c/0x300
...
 [] ? kmalloc_order+0x24/0x90
 [] ? kmalloc_order_trace+0x24/0x220
 [] ? __kmalloc+0x330/0x540
 [] ? sctp_getsockopt_local_addrs+0x174/0xca0 [sctp]
 [] ? sctp_getsockopt+0x10d/0x1b0 [sctp]
 [] ? sock_common_getsockopt+0xb9/0x150
 [] ? SyS_getsockopt+0x1a5/0x270

Signed-off-by: Jiri Slaby 
Cc: Vlad Yasevich 
Cc: Neil Horman 
Cc: "David S. Miller" 
Cc: linux-sctp@vger.kernel.org
Cc: netdev@vger.kernel.org
Acked-by: Neil Horman 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

ipv4: use the right lock for ping_group_range

2016-11-15T06:46:38+00:00

[ Upstream commit 396a30cce15d084b2b1a395aa6d515c3d559c674 ]

This reverts commit a681574c99be23e4d20b769bf0e543239c364af5
("ipv4: disable BH in set_ping_group_range()") because we never
read ping_group_range in BH context (unlike local_port_range).

Then, since we already have a lock for ping_group_range, those
using ip_local_ports.lock for ping_group_range are clearly typos.

We might consider to share a same lock for both ping_group_range
and local_port_range w.r.t. space saving, but that should be for
net-next.

Fixes: a681574c99be ("ipv4: disable BH in set_ping_group_range()")
Fixes: ba6b918ab234 ("ping: move ping_group_range out of CONFIG_SYSCTL")
Cc: Eric Dumazet 
Cc: Eric Salo 
Signed-off-by: Cong Wang 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

ipv4: disable BH in set_ping_group_range()

2016-11-15T06:46:38+00:00

[ Upstream commit a681574c99be23e4d20b769bf0e543239c364af5 ]

In commit 4ee3bd4a8c746 ("ipv4: disable BH when changing ip local port
range") Cong added BH protection in set_local_port_range() but missed
that same fix was needed in set_ping_group_range()

Fixes: b8f1a55639e6 ("udp: Add function to make source port for UDP tunnels")
Signed-off-by: Eric Dumazet 
Reported-by: Eric Salo 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: add recursion limit to GRO

2016-11-15T06:46:38+00:00

[ Upstream commit fcd91dd449867c6bfe56a81cabba76b829fd05cd ]

Currently, GRO can do unlimited recursion through the gro_receive
handlers.  This was fixed for tunneling protocols by limiting tunnel GRO
to one level with encap_mark, but both VLAN and TEB still have this
problem.  Thus, the kernel is vulnerable to a stack overflow, if we
receive a packet composed entirely of VLAN headers.

This patch adds a recursion counter to the GRO layer to prevent stack
overflow.  When a gro_receive function hits the recursion limit, GRO is
aborted for this skb and it is processed normally.  This recursion
counter is put in the GRO CB, but could be turned into a percpu counter
if we run out of space in the CB.

Thanks to Vladimír Beneš  for the initial bug report.

Fixes: CVE-2016-7039
Fixes: 9b174d88c257 ("net: Add Transparent Ethernet Bridging GRO support.")
Fixes: 66e5133f19e9 ("vlan: Add GRO support for non hardware accelerated vlan")
Signed-off-by: Sabrina Dubroca 
Reviewed-by: Jiri Benc 
Acked-by: Hannes Frederic Sowa 
Acked-by: Tom Herbert 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

bridge: multicast: restore perm router ports on multicast enable

2016-11-15T06:46:38+00:00

[ Upstream commit 7cb3f9214dfa443c1ccc2be637dcc6344cc203f0 ]

Satish reported a problem with the perm multicast router ports not getting
reenabled after some series of events, in particular if it happens that the
multicast snooping has been disabled and the port goes to disabled state
then it will be deleted from the router port list, but if it moves into
non-disabled state it will not be re-added because the mcast snooping is
still disabled, and enabling snooping later does nothing.

Here are the steps to reproduce, setup br0 with snooping enabled and eth1
added as a perm router (multicast_router = 2):
1. $ echo 0 > /sys/class/net/br0/bridge/multicast_snooping
2. $ ip l set eth1 down
^ This step deletes the interface from the router list
3. $ ip l set eth1 up
^ This step does not add it again because mcast snooping is disabled
4. $ echo 1 > /sys/class/net/br0/bridge/multicast_snooping
5. $ bridge -d -s mdb show


At this point we have mcast enabled and eth1 as a perm router (value = 2)
but it is not in the router list which is incorrect.

After this change:
1. $ echo 0 > /sys/class/net/br0/bridge/multicast_snooping
2. $ ip l set eth1 down
^ This step deletes the interface from the router list
3. $ ip l set eth1 up
^ This step does not add it again because mcast snooping is disabled
4. $ echo 1 > /sys/class/net/br0/bridge/multicast_snooping
5. $ bridge -d -s mdb show
router ports on br0: eth1

Note: we can directly do br_multicast_enable_port for all because the
querier timer already has checks for the port state and will simply
expire if it's in blocking/disabled. See the comment added by
commit 9aa66382163e7 ("bridge: multicast: add a comment to
br_port_state_selection about blocking state")

Fixes: 561f1103a2b7 ("bridge: Add multicast_snooping sysfs toggle")
Reported-by: Satish Ashok 
Signed-off-by: Nikolay Aleksandrov 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: pktgen: remove rcu locking in pktgen_change_name()

2016-11-15T06:46:38+00:00

[ Upstream commit 9a0b1e8ba4061778897b544afc898de2163382f7 ]

After Jesper commit back in linux-3.18, we trigger a lockdep
splat in proc_create_data() while allocating memory from
pktgen_change_name().

This patch converts t->if_lock to a mutex, since it is now only
used from control path, and adds proper locking to pktgen_change_name()

1) pktgen_thread_lock to protect the outer loop (iterating threads)
2) t->if_lock to protect the inner loop (iterating devices)

Note that before Jesper patch, pktgen_change_name() was lacking proper
protection, but lockdep was not able to detect the problem.

Fixes: 8788370a1d4b ("pktgen: RCU-ify "if_list" to remove lock in next_to_run()")
Reported-by: John Sperbeck 
Signed-off-by: Eric Dumazet 
Cc: Jesper Dangaard Brouer 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman